TNF-.alpha., also known as cachectin, is a 17 kDa (kilodalton) protein produced by cells of the monocyte/macrophage lineage, and by other cells. A variety of biological effects, both beneficial and deleterious, have been attributed to TNF-.alpha.. TNF-.alpha. is beneficial, e.g., in that it is believed to be a part of host anti-tumor defenses. It also produces detrimental effects, however, including, e.g., cardiovascular (shock, ARDS, capillary leakage syndrome), renal (nephritis, acute tubal necrosis), and gastrointestinal (ischemia, colitis, hepatic necrosis) effects, and effects on the central nervous system (fever, anorexia, altered pituitary hormone secretion). In view of the foregoing, a consensus view has developed that TNF-.alpha. is a key mediator of inflammation (including inflammatory diseases such as arthritis) and mammalian responses to injury, invasion by pathogens, and neoplasia.
The biosynthesis of human TNF-.alpha. proceeds by way of a membrane-bound precursor containing 233 amino acid residues [Wang et al., Science 228:149-154 (1985); Muller et al., Nature 335:265-267 (1987)], which is processed during cellular activation by cleavage of a 76-residue peptide to produce the mature, secreted form of TNF-.alpha.. The enzyme(s) responsible for this cleavage, called TNF-.alpha. convertase, has until the present invention been elusive for most mammalian species.
A putative TNF-.alpha. convertase, called PR-3, has been isolated and cloned from human neutrophils, and it has been suggested that this enzyme can be used in screens to identify TNF-.alpha. convertase inhibitors. See International Patent Applications Publication Numbers WO 94/00555 and WO 95/24501. This enzyme, however, is not believed to be the physiologically relevant human TNF-.alpha. convertase because it is a serine protease, whereas the relevant enzyme is believed to be a metalloproteinase. Moreover, the source of the serine protease, neutrophils, is not believed to be important in the production of TNF-.alpha., and the serine protease does not cleave the precursor form of TNF-.alpha. (proTNF-.alpha.) at the point expected for the physiologically relevant human enzyme.
Mohler et al. [Nature 70:218 (1994)] have partially purified another TNF-.alpha. convertase from the human monocytic cell line THP-1. This preparation, however, was very impure, and little could be said about the nature of the TNF-.alpha. convertase in the crude protein mixture of Mohler et al.
In view of the important role of TNF-.alpha. in many disease processes, there is a need for agents that can selectively block the biosynthesis of mature, secreted TNF-.alpha.. The search for such agents would be greatly facilitated by the availability of substantially pure mammalian TNF-.alpha. convertases.